Adopting software engineering trends in AI

Dragan Gašević, Dragan Djurić, Vladan Devedžić

Research output: Contribution to journalArticleResearchpeer-review

9 Citations (Scopus)


The integration of Artificial Intelligence (AI) and software engineering (SE) has led to the development of Air, which is an integrated AI development environment. Air is developed to provide a general modeling and metamodeling infrastructure for AI systems analysis, design, and development. It follows the MDA metamodeling architecture, which provides the means for defining modeling languages. Users can specify the metamodels of modeling languages using the Meta-object Facility (MOF). Air stores its domain models in its model base, which is implemented as a MOF-based repository. In Air framework, XML Schema and XML document are used for sharing MDA metadata. The Air workbench lets users specify the XML-based metamodel of a desired representational format, language or paradigm and put the metamodel in repository. The other implementation of repository is the NetBeans MDR, which is used in Air because of its generic implementation of JMI interfaces.

Original languageEnglish
Pages (from-to)59-66
Number of pages8
JournalIEEE Intelligent Systems
Issue number1
Publication statusPublished - 2007
Externally publishedYes

Cite this